Method for monitoring a production plant and corresponding production plant
By linking the operating status of production equipment with reference values in a storage table, and utilizing machine learning and sensor monitoring, the problem of setting appropriate reference values for coating equipment has been solved, enabling timely fault detection and efficient monitoring.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- DUERR SYSTEMS GMBH
- Filing Date
- 2025-02-03
- Publication Date
- 2026-07-10
AI Technical Summary
Existing methods for monitoring coating equipment are difficult to set appropriate reference values for different operating states, resulting in delayed and time-consuming fault detection. Machine learning methods cannot fully cover all states and lack readability.
By associating different operating states of production equipment with reference values in a storage table, the reference values are determined using supervised or unsupervised machine learning, and real-time monitoring is performed using sensors and monitoring devices, combined with manual input to optimize the reference value settings.
It enables the setting of appropriate reference values for different operating states, improving the timeliness and efficiency of fault detection and simplifying the process of setting reference values.
Smart Images
Figure CN122374716A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to a monitoring method for production equipment (e.g., coating equipment). Furthermore, this invention also relates to a supporting production equipment equipped with a monitoring device for implementing the monitoring method of this invention. Background Technology
[0002] Modern painting equipment used for coating vehicle body parts can operate under different conditions, which differ in various operating parameters and are also known as operating condition categories. For example, one operating condition can be defined by supplying blue paint at a painting pressure of 4.5 bar, while another operating condition can be defined by supplying red paint at a painting pressure of 4.5 bar. Therefore, the paint color (red or blue) and the painting pressure (4.5 bar) constitute the variable operating parameters that define the corresponding operating conditions.
[0003] The known method for monitoring such coating equipment during operation is to measure various operating parameters of the equipment and compare them with specified reference values to detect potential faults. For example, the coating volume can be measured as an operating parameter. Under the first operating condition described above, using blue paint and a coating pressure of 4.5 bar, the coating volume range is 110ml-120ml; while under the second operating condition described above, using red paint and a coating pressure of 4.5 bar, the coating volume range is 130ml-135ml. These extreme values (minimum and maximum) of the coating volume constitute reference values, used for comparison with the measured coating volume as the operating parameter. If the measured value exceeds the specified range, a fault message can be generated.
[0004] The problem with this known monitoring method is that the number of operating parameters that can be set during the operation of coating equipment is numerous, and the number of operating parameters that need to be measured is also extremely large. Therefore, it is difficult to set appropriate reference values for each operating parameter for different operating states.
[0005] On the one hand, this often leads to overly coarse reference values being set in actual operation, as accurately setting reference values for all operating states requires a huge amount of work. In practical applications, this means that potential faults can only be detected later.
[0006] On the other hand, determining appropriate reference values for numerous operating states can take a lot of time.
[0007] One approach to solving this problem is to employ machine learning methods (“machine learning”), such as the technique disclosed in DE102019112099B3. In this approach, the reference value is automatically determined through machine learning. However, this approach still has many problems.
[0008] For example, machine learning can hardly cover all possible different operating states.
[0009] Furthermore, supervised learning requires identifying so-called labels that indicate the corresponding operational status, but such labels are often unavailable in practical applications.
[0010] Furthermore, neural networks commonly used in machine learning lack human readability, making it impossible for operators to determine whether a certain operating state has been fully represented by the underlying model.
[0011] Finally, for the general technical background of the present invention, reference may also be made to DE102009013561A1, DE102015112361A1, WO2006 / 037137A1 and WO84 / 02592A1. Summary of the Invention
[0012] Therefore, the object of the present invention is to provide an improved method for monitoring production equipment. Furthermore, another object of the present invention is to provide a production equipment capable of implementing the monitoring method of the present invention.
[0013] The monitoring method of the present invention is generally applicable to production equipment, and is preferably applied to coating equipment that coats components (e.g., motor vehicle body components) with a coating agent (e.g., paint).
[0014] Consistent with the prior art described above, the monitoring method of the present invention first requires setting the target operating state of the production equipment (e.g., painting equipment). For example, a target paint (e.g., red paint) and a target painting pressure (e.g., 4.5 bar) for painting automotive body parts can be set, and this target operating state is defined by these two operating parameters. However, the present invention is not limited to these two operating parameters (paint color and painting pressure) in defining the operating state; on the contrary, the corresponding operating state of the production equipment can also be determined by other operating parameters, in particular by multiple operating parameters, which will be described in detail below.
[0015] In addition, consistent with the prior art described above, the monitoring method of the present invention also includes: making the production equipment operate according to the previously set target operating state.
[0016] During the operation of the production equipment according to the previously set target operating state, the operating parameters of the production equipment are measured. For example, the amount of coating applied, as described in the prior art section above, can be measured, i.e., the amount of paint consumed for painting the vehicle body.
[0017] Furthermore, consistent with the prior art described above, the monitoring method of the present invention further includes: monitoring the production equipment by comparing the measured operating parameters (e.g., the applied coating amount) with reference values (e.g., maximum and minimum values) of the corresponding operating parameters. Based on the comparison results, for example, when the measured value of the operating parameter exceeds a specified allowable range, fault information can be generated.
[0018] The difference between this invention and the prior art is the method for determining the reference value used for comparison with measured operating parameters.
[0019] First, for the various operating states of the production equipment, determine the reference values corresponding to the operating parameters (such as coating amount) based on the corresponding operating state of the production equipment (e.g., limited by paint color and coating pressure).
[0020] Subsequently, the determined reference values are associated with the corresponding operating states and stored in a storage table (database). Therefore, this storage table can store monitoring reference values corresponding to a large number of potential operating states of production equipment.
[0021] During the operation of production equipment, the corresponding reference value can be conveniently read from the storage table according to the current operating status of the production equipment.
[0022] This method can set appropriate reference values for a large number of operating states of production equipment, thereby avoiding the various problems mentioned above.
[0023] In one embodiment of the present invention, the reference value corresponding to the corresponding operating state is determined from historical measurement data of the operating parameters through supervised or unsupervised machine learning.
[0024] Alternatively, in a simplified embodiment, the reference value corresponding to the corresponding operating state can be determined by calculating the average value of each operating parameter when the production equipment is running in the corresponding operating state.
[0025] For example, the above average can be calculated based on a specific operating cycle or a specific production batch (in particular, a specific number of painted vehicle bodies). Furthermore, the average is preferably calculated using a rolling method, particularly based on the immediately preceding operating cycle or the immediately preceding batches (e.g., the immediately preceding batches of painted vehicle bodies).
[0026] The reference values corresponding to the operating states of production equipment can also be determined in the following ways: First, the production equipment is operated in the corresponding operating state; then, the fault-free operating cycle or a single fault-free event (e.g., a car body completed without fault) is determined when the production equipment is operating in that operating state; next, operating parameters (e.g., coating amount) are measured during the fault-free operating cycle or the fault-free event; finally, the reference values are determined based on the operating parameters measured during the fault-free operating cycle or the fault-free event.
[0027] Alternatively, reference values for operating parameters can be set manually by the operator.
[0028] As mentioned earlier, the operating status of production equipment can be determined by paint color and coating pressure. However, within the scope of the monitoring method of this invention, the operating status of production equipment can also be determined by various other operating parameters, as shown in the following examples: - The type of paint to be applied; - The pressure of the coating to be applied; - The mass flow rate of the coating to be applied; - The rotational speed of the rotary atomizer used to apply the coating; - Charging voltage and / or charging current of the electrostatic coating charging system; - The shaping airflow used to form the jet stream of the rotary atomizer; - Robot motion path, especially the coating path at the coating application end of a path-controlled coating robot; - Coating agent temperature; - Coating viscosity; - Coating valve opening and closing time.
[0029] The corresponding operating state of production equipment can be determined by multiple operating parameters, which can be arbitrarily combined to define the corresponding operating state. Therefore, the present invention is not limited to the example mentioned above where the operating state is defined by the two operating parameters of paint color and coating pressure; the operating state of production equipment can be defined by two or more, three, four, five or six different operating parameters.
[0030] As mentioned earlier, the coating amount can be measured as an operating parameter to be monitored in coating equipment. However, within the scope of the monitoring method of this invention, other operating parameters of production equipment can also be measured and monitored, as shown in the following examples: - Paint flow rate; - Coating pressure of the coating agent; - Charging voltage and / or charging current of the electrostatic coating charging system; - Forming air pressure and / or forming air flow rate used to form the paint jet stream; - The rotational speed of the rotary atomizer used to apply the coating; - The driving air pressure that drives the rotating atomizer turbine; - The torque and / or current of the drive motor for the robot's drive axis.
[0031] The preceding text only used the monitoring of a single operating parameter, the actual coating amount, in coating equipment as an example to illustrate the present invention, but the monitoring method of the present invention is also applicable to the monitoring of multiple operating parameters of production equipment.
[0032] As mentioned earlier, the storage table stores reference values for the operating parameters (e.g., coating amount) to be monitored corresponding to different operating states (e.g., different paint colors and coating pressures). Within the scope of this invention, the relationship between the operating state and the corresponding reference value of the operating parameter is defined by so-called rules. In the simplest case, each operating parameter to be monitored corresponds to a target reference value and an allowable tolerance; exceeding the tolerance range is determined to be a fault state; however, the rules can also be more complex.
[0033] Within the scope of this invention, the above rules can be defined in various ways, which are briefly described below: - Unsupervised learning: For example, calculating a rolling average based on historical data and using it as a reference value, or employing more complex computational logic.
[0034] - Supervised learning: Operators are limited to normal operating conditions (i.e., fault-free and qualified operating conditions, such as a specific operating cycle or a single qualified workpiece), and reference values are calculated only based on data from such normal operating conditions.
[0035] - User / expert manual input: In particular, reference values for special operating conditions that are difficult to learn automatically in actual operation but have been determined through experience or laboratory testing can be entered.
[0036] It should also be noted that, in the context of this invention, the term "operating state" can further define a segment ("path segment") of the coating path traveled by the coating application end of the coating robot. Each path segment can be configured with different operating parameters (e.g., paint flow rate, forming air flow rate, rotary atomizer speed, electrostatic coating charging system charging voltage, etc.), where these operating parameters characterize the operating state of the corresponding segment of the coating path. Therefore, during the travel of the coating path, the operating parameters of the coating equipment can be dynamically adjusted segment by segment; simultaneously, specific reference values can be set for each segment of the coating path and its corresponding operating state.
[0037] In addition to the monitoring methods described above, this invention also claims protection for a production apparatus designed to perform the monitoring methods of this invention. For example, the production apparatus may be a coating apparatus that coats a component (e.g., a motor vehicle body component) with a coating agent (e.g., paint), but this invention is not limited to coating apparatus in terms of the type of production apparatus.
[0038] The production equipment of the present invention is first equipped with multiple sensors for measuring operating parameters of the production equipment (e.g., paint color and coating pressure) and / or operating parameters to be monitored (e.g., coating amount) during equipment operation. For example, the measurable operating parameter in the coating equipment is the applied coating amount.
[0039] In addition, consistent with conventional production equipment, the production equipment of the present invention is equipped with a control system for setting the target operating state of the production equipment and querying measured operating parameters (e.g., coating amount) and / or operating status parameters (e.g., paint color and coating pressure) from sensors.
[0040] Furthermore, consistent with known production equipment, the production equipment of the present invention also includes a monitoring device for monitoring the production equipment by comparing the operating parameters of the production equipment measured by the sensors with preset operating parameter reference values; based on the comparison results, fault information can be generated or other response actions can be triggered.
[0041] The difference between the production equipment of the present invention and the prior art is that the monitoring device has a storage table (database) for storing reference values of operating parameters in association with the corresponding operating status.
[0042] Furthermore, the monitoring device preferably includes a rule generator for determining reference values for operating parameters based on measured operating parameters. The relationship between the operating parameters of the production equipment used to define the operating state (e.g., paint color and coating pressure) and the corresponding reference values of the operating parameters to be monitored (e.g., coating amount reference values) constitutes the rules generated by the rule generator.
[0043] The rule generator can determine the reference value corresponding to the corresponding operating state from historical measurement data of operating parameters through supervised or unsupervised machine learning.
[0044] Alternatively, the rule generator can calculate the reference value as the average of the measured operating parameters, or directly use the measured operating parameters during a fault-free operating cycle or a fault-free event process as the reference value.
[0045] In addition, the monitoring device may have a rule editor for operators to manually input reference values.
[0046] Other preferred embodiments of the present invention are described in the dependent claims, or are explained in conjunction with the description of the preferred embodiments of the present invention in conjunction with the accompanying drawings. Attached Figure Description
[0047] Figure 1 The flowchart illustrates the monitoring method of the present invention applied to the painting equipment for motor vehicle body parts.
[0048] Figure 2 The flowchart illustrates the process of determining reference values by calculating the rolling average of measured operating parameters.
[0049] Figure 3 for Figure 2 An improved implementation wherein the reference value is determined by measured operating parameters during a fault-free operating cycle or during a fault-free event process.
[0050] Figure 4 This is a schematic diagram of the coating equipment of the present invention.
[0051] Figure 5 This is a schematic diagram of the storage table structure of the present invention, showing the correspondence between reference values and different operating states of the coating equipment.
[0052] Figure 6 This is a schematic diagram of the coating production line, which is divided into multiple sections. Each section is set with a certain operating state and a certain operating parameter is monitored. Detailed Implementation
[0053] The following text combines Figure 1 The flowchart shown illustrates the monitoring method of the present invention using a coating equipment as an example.
[0054] In the first step S1, the target operating state of the coating equipment is selected. In this example, the operating state of the coating equipment is limited to applying red paint at a coating pressure of 4.5 bar. However, this is a simplified example and is only used to help understand the present invention; in actual applications, the operating state of the coating equipment can be limited by a variety of different operating parameters.
[0055] In the second step S2, reference values corresponding to the coating equipment operating parameters are read from the storage table according to the operating state selected in step S1. In this example, the reference value read is the applied coating amount, which is retrieved from the storage table according to the previously selected operating state. For example, it may represent the maximum and minimum applied coating amount.
[0056] In the next step S3, the coating equipment is operated according to the selected operating state, that is, in this example, red paint is applied at a coating pressure of 4.5 bar.
[0057] In step S4, during the operation of the coating equipment in the selected operating state, the operating parameters of the coating equipment are measured. In this simplified example, the measured operating parameter is the applied coating amount.
[0058] In the next step S5, the coating equipment is monitored by comparing the measured operating parameters (coating amount) with the corresponding reference values. For example, ... Figure 5 As shown, when applying blue paint at a coating pressure of 4.5 bar, the reference value for the amount of paint applied is 120 ml, with a tolerance of 5 ml.
[0059] In step S6, if the measured operating parameters (e.g., coating amount) exceed the allowable range set by the reference value (e.g., 120ml±5ml), a fault signal is output; at the same time, the control system can trigger corresponding countermeasures accordingly.
[0060] The following text combines Figure 2The flowchart shown illustrates how the reference value is determined.
[0061] In step S1, the target operating state of the coating equipment is first selected, for example, applying red paint with a coating pressure of 4.5 bar.
[0062] In the next step S2, the coating equipment is operated according to the selected operating state.
[0063] In step S3, operating parameters (e.g., the amount of coating applied) are measured during the operation of the coating equipment.
[0064] In step S4, the rolling average of the measured operating parameters (e.g., the amount of coating applied) of the coating equipment when it is running in the selected operating state is calculated.
[0065] In the next step S5, the reference value corresponding to the operating parameter is determined based on the previously calculated rolling average value of the measured operating parameters (e.g., the amount of coating applied).
[0066] The following text combines Figure 3 The flowchart shown illustrates another alternative embodiment of the reference value determination method. To avoid repetition, please refer to the previous section on... Figure 2 Explanation.
[0067] The feature of this embodiment is that during the operation of the painting equipment in the target operating state, a fault-free operating cycle or a fault-free event (such as a car body that has been painted without fault) is determined; then, a monitoring reference value is determined based on the measured operating parameters during the fault-free operating cycle or the fault-free event.
[0068] Figure 4 This is a schematic diagram of the structure of the production equipment 1 of the present invention. The production equipment can be the painting equipment mentioned above for painting motor vehicle body parts.
[0069] Production equipment 1 is connected to control system 2 via an interface. On the one hand, control system 2 can control production equipment 1 through the interface and specify the target operating state by setting operating parameters (such as paint color and coating pressure); on the other hand, control system 2 can read operating parameters (such as coating amount) measured by sensors in production equipment 1 through the interface.
[0070] The control system 2 can first store the measured operating parameters (such as coating amount) and the set operating status parameters (such as paint color and coating pressure) as raw data in the database 3.
[0071] The raw data stored in database 3 is processed and then stored in another database 4 as processed data.
[0072] In addition, there is a storage table 5, which stores reference values (e.g., maximum and minimum values) corresponding to the operating parameters (e.g., coating amount) to be monitored under different operating states (e.g., different paint colors and coating pressures) of production equipment 1.
[0073] The reference values used to monitor production equipment 1 (such as the maximum and minimum coating amounts) are all stored in storage table 5 in association with the corresponding operating status; therefore, storage table 5 contains a large number of adaptive reference values corresponding to different operating statuses.
[0074] Reference values can be generated by the automatic rule generator 6: the rule generator 6 retrieves the processed raw data from the database 4 and stores the reference values in the storage table 5 in association with the corresponding running status.
[0075] Alternatively, reference values can also be generated by the manual rule editor 7 and stored in the storage table 5; the manual rule editor 7 can interact with the operator 9 through the human-machine interface 8.
[0076] In addition, a monitoring module 10 is provided to execute the previously generated rules and send control signals to the control system 2.
[0077] Figure 5 The table is a simplified diagram showing four different operating states, represented by different paint colors and different coating pressures, with each row corresponding to one operating state. At the same time, the table shows the reference value of the operating parameter (coating amount) to be monitored for each operating state (each row).
[0078] at last, Figure 6 This is a schematic diagram of the coating path 11. The coating application end of the coating equipment (e.g., a rotary atomizer) is guided by a multi-axis coating robot to travel along the coating path 11 and coat the surface of the target part. The coating path 11 is defined by multiple path points P1-P8, which are determined according to the geometry of the part to be coated; this process is called teaching. The coating path 11 consists of multiple continuous path segments BA1-BA7. During the movement of the coating path 11, the operating state of the coating equipment can be dynamically adjusted in each path segment BA1-BA7. Therefore, different operating states can be set for each path segment BA1-BA7, and for each path segment BA1-BA7 and its corresponding operating state, the reference values corresponding to the operating parameters to be monitored can be read from a storage table.
[0079] This invention is not limited to the preferred embodiments described above; on the contrary, various modifications and variations can be made without departing from the core concept of this invention, all of which fall within the scope of protection of this invention. In particular, this invention independently claims protection for the subject matter and technical features described in the dependent claims, without referencing the corresponding claims, and especially without referencing the technical features of the independent claims; therefore, this invention comprises multiple independently protected technical solutions.
[0080] List of reference numerals 1. Production equipment 2. Control system with interface 3. Database for storing raw measured data 4. Database for storing processed raw data 5. Storage table, containing running status and corresponding reference values. 6. Automatic Rule Generator 7. Manual Rule Editor 8. Human-computer interface 9. Operators 10 Monitoring Module 11 Painting Path BA1-BA7 painting path segment P1-P8 Painting Path Points
Claims
1. A monitoring method for a production apparatus (1), particularly a coating apparatus for coating parts with a coating agent, and particularly a coating apparatus for coating motor vehicle body parts with a paint, comprising the following steps: a) Set the target operating status of the production equipment (1), in particular by setting the target paint and target coating pressure for painting motor vehicle body parts; b) Make the production equipment (1) operate according to the set operating state; c) During the operation of the production equipment (1) in the set operating state, measure the operating parameters of the production equipment (1), especially by measuring the amount of coating applied; d) Monitor the production equipment by comparing the determined operating parameters with the reference values of the operating parameters (1). The monitoring method is characterized by comprising the following steps for determining a reference value: e) For the various operating states of the production equipment (1), determine the reference values of the operating parameters according to the corresponding operating states of the production equipment (1); f) Associate the determined reference values with the corresponding operating states and store them in storage table (5); and g) Read the reference value from the storage table (5) based on the current operating status of the production equipment (1).
2. The monitoring method according to claim 1, characterized in that, The reference values for the corresponding operating status are determined from historical measurement data of the operating parameters through supervised or unsupervised machine learning.
3. The monitoring method according to claim 1, characterized in that, The reference values for each operating state are determined by calculating the average value of the corresponding operating parameters of the production equipment (1) when it is running in the corresponding operating state.
4. The monitoring method according to claim 3, characterized in that, a) The average value is calculated based on a specific operating cycle or a specific production batch, particularly a specific number of painted vehicle bodies; and / or b) Averages are calculated using a rolling method, particularly based on the immediate preceding operating cycle or the immediate preceding batches.
5. The monitoring method according to any one of the preceding claims, characterized in that, The monitoring method includes the following steps for determining reference values for the corresponding operating status of the production equipment (1): a) To enable the production equipment (1) to operate in the corresponding operating state; b) Determine the fault-free operating cycle or fault-free events of the production equipment (1) under the corresponding operating conditions, especially the fault-free painting process of motor vehicle bodies; c) Measure operating parameters during a trouble-free operating cycle or during a trouble-free event; d) Set reference values based on the measured operating parameters during the fault-free operation cycle or during the fault-free event process.
6. The monitoring method according to any one of the preceding claims, characterized in that, The reference values for the operating parameters are set by the operator.
7. The monitoring method according to any one of the preceding claims, characterized in that, a) The operating status of production equipment (1) is determined by at least one of the following variables: a1) The type of paint to be applied; a2) The pressure of the coating agent to be applied; a3) The mass flow rate of the coating agent to be applied; a4) Rotational speed of the rotary atomizer used to apply the coating agent; a5) The charging voltage and / or charging current of the electrostatic coating charging system; a6) The shaping airflow used to form the jet stream of the rotary atomizer; a7) Robot motion path, especially the coating path of the coating application end of a path-controlled coating robot; a8) Coating agent temperature; a9) Coating viscosity; a10) Coating valve opening and closing time, and / or b) The measured operating parameters of the production equipment (1) include at least one of the following variables: b1) Coating coverage; b2) Coating pressure of the coating agent; b3) The charging voltage and / or charging current of the electrostatic coating charging system; b4) Forming air pressure and / or forming air flow rate used to form the paint jet stream; b5) Rotational speed of the rotary atomizer used to apply the coating; b6) The air pressure of the driving air that drives the rotary atomizer turbine; b7) The torque and / or current of the drive motor of the robot drive axis.
8. The monitoring method according to any one of the preceding claims, characterized in that, a) The production equipment (1) is a coating equipment, equipped with a coating robot. The coating robot drives the application device to move along the predetermined coating path (11) on the surface of the component to be coated. b) The coating path (11) is divided into multiple continuous path segments (BA1-BA7). c) For each of the path segments (BA1-BA7) of the coating path (11), the operating state of the coating equipment is defined, wherein the operating state is determined by the operating parameters of the coating equipment; d) For each of the path segments (BA1-BA7) in the coating path (11), determine the reference values for monitoring the operating parameters of the coating equipment; and e) For each of the path segments (BA1-BA7) in the coating path (11), monitor the operating parameters based on the reference values.
9. A production apparatus (1), particularly a coating apparatus for coating parts, particularly a coating apparatus for coating automotive body parts, comprising: a) Multiple sensors are used to measure the operating parameters of the production equipment (1); b) Control system (2), used to set the target operating state of production equipment (1) and query the measured operating parameters from the sensors; as well as c) Monitoring device (5-10) is used to monitor the production equipment (1) by comparing the operating parameters of the production equipment (1) measured by the sensor with the reference values of the operating parameters. Its features are, d) The monitoring device (5-10) is equipped with a storage table (5) for storing reference values of operating parameters in association with corresponding operating states.
10. The production equipment (1) according to claim 9, characterized in that, The monitoring device has a rule generator (6, 7) to determine reference values for operating parameters based on measured operating parameters.
11. The production equipment (1) according to claim 10, characterized in that, The rule generator (6) determines the reference values of the corresponding operating state from the historical measurement data of the operating parameters through supervised or unsupervised machine learning.
12. The production equipment (1) according to claim 10, characterized in that, a) The rule generator (6) calculates the reference value as the average value of the measured operating parameters; or b) The rule generator (6) uses the measured operating parameters during the fault-free operation cycle or the fault-free event process as reference values.
13. The production equipment (1) according to any one of claims 9 to 12, characterized in that, The monitoring device has a rule editor (7) for operators to input reference values.